A new Bayesian method to identify the environmental factors that influence recent migration

We present a new multilocus genotype method that makes inferences about recent immigration rates and identifies the environmental factors that are more likely to explain observed gene flow patterns. It also estimates population-specific inbreeding coefficients, allele frequencies, and local population F(ST)'s and performs individual assignments. We generate synthetic data sets to determine the region of the parameter space where our method is and is not able to provide accurate estimates. Our simulation study indicates that reliable results can be obtained when the global level of genetic differentiation (F(ST)) is >1%, the number of loci is only 10, and sample sizes are of the order of 50 individuals per population. We illustrate our method by applying it to Pakistani human data, considering altitude and geographic distance as explanatory factors. Our results suggest that altitude explains better the genetic data than geographic distance. Additionally, they show that southern low-altitude populations have higher migration rates than northern high-altitude ones.     

Glacial survival and local adaptation in an alpine leaf beetle

The challenge in defining conservation units so that they represent evolutionary entities has been to combine both genetic properties and ecological significance. Here we make use of the complexity of the European Alps, with their genetic landscape shaped by geographical barriers and postglacial colonization, to examine the correlation between ecological and genetic divergence. Montane species, because of the fragmentation of their present habitat, constitute extreme cases in which to test if genetically distinct subgroups based on neutral markers are also ecologically differentiated and show local adaptation. In the leaf beetle Oreina elongata, populations show variation in host plant use and a patchy distribution throughout the Alps and Apennines. We demonstrate that despite very strong genetic isolation (F(ST) = 0.381), variation in host plant use has led to differences in larval life-history traits between populations only as a secondary effect of host defence chemistry, and not through physiological adaptation to plant nutritional value. We also establish that populations that are more ecologically different in terms of larval performance are also more genetically divergent. In addition, morphological variation used to define subspecies appears to be mirrored in the population genetics of this species, resulting in almost perfect clustering based on microsatellite data. Finally, we argue from their strong genetic structure and congruent distribution that the subspecies of O. elongata were divided among the same glacial refugia within the Alps that have been proposed for alpine plants.     

Microsatellite allele sizes: a simple test to assess their significance on genetic differentiation

The mutation process at microsatellite loci typically occurs at high rates and with stepwise changes in allele sizes, features that may introduce bias when using classical measures of population differentiation based on allele identity (e.g., F(ST), Nei's Ds genetic distance). Allele size-based measures of differentiation, assuming a stepwise mutation process [e.g., Slatkin's R(ST), Goldstein et al.'s (deltamu)(2)], may better reflect differentiation at microsatellite loci, but they suffer high sampling variance. The relative efficiency of allele size- vs. allele identity-based statistics depends on the relative contributions of mutations vs. drift to population differentiation. We present a simple test based on a randomization procedure of allele sizes to determine whether stepwise-like mutations contributed to genetic differentiation. This test can be applied to any microsatellite data set designed to assess population differentiation and can be interpreted as testing whether F(ST) = R(ST). Computer simulations show that the test efficiently identifies which of F(ST) or R(ST) estimates has the lowest mean square error. A significant test, implying that R(ST) performs better than F(ST), is obtained when the mutation rate, mu, for a stepwise mutation process is (a) >/= m in an island model (m being the migration rate among populations) or (b) >/= 1/t in the case of isolated populations (t being the number of generations since population divergence). The test also informs on the efficiency of other statistics used in phylogenetical reconstruction [e.g., Ds and (deltamu)(2)], a nonsignificant test meaning that allele identity-based statistics perform better than allele size-based ones. This test can also provide insights into the evolutionary history of populations, revealing, for example, phylogeographic patterns, as illustrated by applying it on three published data sets.     

Empirical Bayes inference of pairwise F(ST) and its distribution in the genome

Populations often have very complex hierarchical structure. Therefore, it is crucial in genetic monitoring and conservation biology to have a reliable estimate of the pattern of population subdivision. F(ST)'s for pairs of sampled localities or subpopulations are crucial statistics for the exploratory analysis of population structures, such as cluster analysis and multidimensional scaling. However, the estimation of F(ST) is not precise enough to reliably estimate the population structure and the extent of heterogeneity. This article proposes an empirical Bayes procedure to estimate locus-specific pairwise F(ST)'s. The posterior mean of the pairwise F(ST) can be interpreted as a shrinkage estimator, which reduces the variance of conventional estimators largely at the expense of a small bias. The global F(ST) of a population generally varies among loci in the genome. Our maximum-likelihood estimates of global F(ST)'s can be used as sufficient statistics to estimate the distribution of F(ST) in the genome. We demonstrate the efficacy and robustness of our model by simulation and by an analysis of the microsatellite allele frequencies of the Pacific herring. The heterogeneity of the global F(ST) in the genome is discussed on the basis of the estimated distribution of the global F(ST) for the herring and examples of human single nucleotide polymorphisms (SNPs).     

Microsatellite null alleles and estimation of population differentiation

Microsatellite null alleles are commonly encountered in population genetics studies, yet little is known about their impact on the estimation of population differentiation. Computer simulations based on the coalescent were used to investigate the evolutionary dynamics of null alleles, their impact on F(ST) and genetic distances, and the efficiency of estimators of null allele frequency. Further, we explored how the existing method for correcting genotype data for null alleles performed in estimating F(ST) and genetic distances, and we compared this method with a new method proposed here (for F(ST) only). Null alleles were likely to be encountered in populations with a large effective size, with an unusually high mutation rate in the flanking regions, and that have diverged from the population from which the cloned allele state was drawn and the primers designed. When populations were significantly differentiated, F(ST) and genetic distances were overestimated in the presence of null alleles. Frequency of null alleles was estimated precisely with the algorithm presented in Dempster et al. (1977). The conventional method for correcting genotype data for null alleles did not provide an accurate estimate of F(ST) and genetic distances. However, the use of the genetic distance of Cavalli-Sforza and Edwards (1967) corrected by the conventional method gave better estimates than those obtained without correction. F(ST) estimation from corrected genotype frequencies performed well when restricted to visible allele sizes. Both the proposed method and the traditional correction method have been implemented in a program that is available free of charge at http://www.montpellier.inra.fr/URLB/. We used 2 published microsatellite data sets based on original and redesigned pairs of primers to empirically confirm our simulation results.     

Spatial and temporal determinants of genetic structure in Gentianella bohemica

The biennial plant Gentianella bohemica is a subendemic of the Bohemian Massif, where it occurs in seminatural grasslands. It has become rare in recent decades as a result of profound changes in land use. Using amplified fragment length polymorphisms (AFLP) fingerprint data, we investigated the genetic structure within and among populations of G. bohemica in Bavaria, the Czech Republic, and the Austrian border region. The aim of our study was (1) to analyze the genetic structure among populations and to discuss these findings in the context of present and historical patterns of connectivity and isolation of populations, (2) to analyze genetic structure among consecutive generations (cohorts of two consecutive years), and (3) to investigate relationships between intrapopulational diversity and effective population size (N(e)) as well as plant traits. (1) The German populations were strongly isolated from each other (pairwise F(ST)= 0.29-0.60) and from all other populations (F(ST)= 0.24-0.49). We found a pattern of near panmixis among the latter (F(ST)= 0.15-0.35) with geographical distance explaining only 8% of the genetic variance. These results were congruent with a principal coordinate analysis (PCoA) and analysis using STRUCTURE to identify genetically coherent groups. These findings are in line with the strong physical barrier and historical constraints, resulting in separation of the German populations from the others. (2) We found pronounced genetic differences between consecutive cohorts of the German populations (pairwise F(ST)= 0.23 and 0.31), which can be explained by local population history (land use, disturbance). (3) Genetic diversity within populations (Shannon index, H(Sh)) was significantly correlated with N(e) (R(S)= 0.733) and reflected a loss of diversity due to several demographic bottlenecks. Overall, we found that the genetic structure in G. bohemica is strongly influenced by historical periods of high connectivity and isolation as well as by marked demographic fluctuations in declining populations.     

Population structure of a parasitic plant and its perennial host

Characterization of host and parasite population genetic structure and estimation of gene flow among populations are essential for the understanding of parasite local adaptation and coevolutionary interactions between hosts and parasites. We examined two aspects of population structure in a parasitic plant, the greater dodder (Cuscuta europaea) and its host plant, the stinging nettle (Urtica dioica), using allozyme data from 12 host and eight parasite populations. First, we examined whether hosts exposed to parasitism in the past contain higher levels of genetic variation. Second, we examined whether host and parasite populations differ in terms of population structure and if their population structures are correlated. There was no evidence that host populations differed in terms of gene diversity or heterozygosity according to their history of parasitism. Host populations were genetically more differentiated (F(ST) = 0.032) than parasite populations (F(ST) = 0.009). Based on these F(ST) values, gene flow was high for both host and parasite. Such high levels of gene flow could counteract selection for local adaptation of the parasite. We found no significant correlation between geographic and genetic distance (estimated as pairwise F(ST)), either for the host or for the parasite. Furthermore, host and parasite genetic distance matrices were uncorrelated, suggesting that sites with genetically similar host populations are unlikely to have genetically similar parasite populations.     

Microsatellite analysis reveals interpopulation differentiation and gene flow in the endangered tree Changiostyrax dolichocarpa (Styracaceae) with fragmented distribution in central China

Polymorphic simple sequence repeat (SSR) markers were used to investigate the impact of habitat fragmentation on the population structure and gene flow of Changiostyrax dolichocarpa, a critically endangered tree in central China. Intrapopulation genetic diversity, population structure and gene flow in the five extant populations of this species were analysed by eight SSR markers. Intrapopulation genetic diversity results suggest that C. dolichocarpa remnants maintained a relatively high degree of genetic diversity despite severe fragmentation. Low genetic differentiation among populations was found based on Wright's F(ST) and amova analysis. Both the F(ST)-based estimate and private allele method revealed high historical gene flow among the remnant populations. Recent immigrants, detected by assignment tests, tend to decrease from the grandparent generation to the current generation. The potentially highly restricted current gene flow among fragments may render the fragmented populations of C. dolichocarpa at a higher risk of local extinction several generations after fragmentation. Both in situ and ex situ conservation management for the remnant populations of C. dolichocarpa are therefore urgently needed to rescue remaining genetic diversity.     

Multilevel analyses of genetic differentiation in Anopheles gambiae s.s. reveal patterns of gene flow important for malaria-fighting mosquito projects

Malaria control projects based on the introduction and spread of transgenes into mosquito populations depend on the extent of isolation between those populations. On the basis of the distribution of paracentric inversions, Anopheles gambiae has been subdivided into five subspecific chromosomal forms. Estimating gene flow between and within these forms of An. gambiae presents a number of challenges. We compared patterns of genetic divergence (F(ST)) between sympatric populations of the Bamako and Mopti forms at five sites. We used microsatellite loci within the j inversion on chromosome 2, which is fixed in the Bamako form but absent in the Mopti form, and microsatellites on chromosome 3, a region void of inversions. Estimates of genetic diversity and F(ST)'s suggest genetic exchanges between forms for the third chromosome but little for the j inversion. These results suggest a role for the inversion in speciation. Extensive gene flow within forms among sites resulted in populations clustering according to form despite substantial gene flow between forms. These patterns underscore the low levels of current gene flow between chromosomal forms in this area of sympatry. Introducing refractoriness genes in areas of the genome void of inversions may facilitate their spread within forms but their passage between forms may prove more difficult than previously thought.     

Analysis of microsatellite variation in Pinus radiata reveals effects of genetic drift but no recent bottlenecks

Most conifer species occur in large continuous populations, but radiata pine, Pinus radiata, occurs only in five disjunctive natural populations in California and Mexico. The Mexican island populations were presumably colonized from the mainland millions of years ago. According to Axelrod (1981), the mainland populations are relicts of an earlier much wider distribution, reduced some 8,000 years ago, whereas according to Millar (1997, 2000), the patchy metapopulation-like structure is typical of the long-term population demography of the species. We used 19 highly polymorphic microsatellite loci to describe population structure and to search for signs of the dynamics of population demography over space and time. Frequencies of null alleles at microsatellite loci were estimated using an approach based on the probability of identity by descent. Microsatellite genetic diversities were high in all populations [expected heterozygosity (H(e)) = 0.68-0.77], but the island populations had significantly lower estimates. Variation between loci in genetic differentiation (F(ST)) was high, but no locus deviated statistically significantly from the rest at an experiment wide level of 0.05. Thus, all loci were included in subsequent analysis. The average differentiation was measured as F(ST) = 0.14 (SD 0.012), comparable with earlier allozyme results. The island populations were more diverged from the other populations and from an inferred common ancestral gene pool than the mainland ones. All populations showed a deficiency of expected heterozygosity given the number of alleles, the mainland populations more so than the island ones. The results thus do not support a recent important contraction in the mainland range of radiata pine.     

Hierarchical Analysis of Nucleotide Diversity in Geographically Structured Populations

Existing methods for analyzing nucleotide diversity require investigators to identify relevant hierarchical levels before beginning the analysis. We describe a method that partitions diversity into hierarchical components while allowing any structure present in the data to emerge naturally. We present an unbiased version of NEI's nucleotide diversity statistics and show that our modification has the same properties as WRIGHT's F(ST). We compare its statistical properties with several other F(ST) estimators, and we describe how to use these statistics to produce a rooted tree of relationships among the sampled populations in which the mean time to coalescence of haplotypes drawn from populations belonging to the same node is smaller than the mean time to coalescence of haplotypes drawn from populations belonging to different nodes. We illustrate the method by applying it to data from a recent survey of restriction site variation in the chloroplast genome of Coreopsis grandiflora.     

Landscape-level spatial genetic structure in Quercus acutissima (Fagaceae)

Quercus acutissima (Fagaceae), a deciduous broad-leaved tree, is an important forest element in hillsides of South Korea. We used allozyme loci, Wright's F statistics, and multilocus spatial autocorrelation statistics to examine the distribution of genetic diversity within and among three local populations and the spatial genetic structure at a landscape scale (15 ha, 250 × 600 m) on Oenaro Island, South Korea. Levels of genetic diversity in Q. acutissima populations were comparable to mean values for other oak species. A moderate but significant deficit of heterozygotes (mean F(IS) = 0.069) was detected within local populations and low but significant differentiation was observed among populations (F(ST) = 0.010). Spatial autocorrelation analyses revealed little evidence of significant genetic structure at spatial scales of 100-120 m. The failure to detect genetic structure within populations may be due to intraspecific competition or random mortality among saplings, resulting in extensive thinning within maternal half-sib groups. Alternatively, low genetic differentiation at the landscape scale indicates substantial gene flow among local populations. Although wind-borne pollen may be the primary source of gene flow in Q. acutissima, these results suggest that acorn movement by animals may be more extensive than previously anticipated. Comparison of these genetic data for Oenaro Island with a disturbed isolated inland population suggests that population-to-population differences in internal genetic structure may be influenced by local variation in regeneration environment (e.g., disturbance).     

Past climate changes explain the phylogeography of Vitellaria paradoxa over Africa

The evolution of the savanna biome has been deeply marked by repeated contraction/expansion phases due to climate perturbations during the Quaternary period. In this study, we investigated the impact of the last glacial maximum (LGM) on the present genetic pattern of Vitellaria paradoxa (shea tree), a major African savanna tree. A range-wide sampling of the species enabled us to sample 374 individuals from 71 populations distributed throughout sub-Sahelian Africa. Trees were genotyped using 3 chloroplasts and 12 nuclear microsatellites, and were sequenced for 2 polymorphic chloroplast intergenic spacers. Analyses of genetic diversity and structure were based on frequency-based and Bayesian methods. Potential distributions of V. paradoxa at present, during the LGM and the last interglacial period, were examined using DIVA-GIS ecological niche modelling (ENM). Haplotypic and allelic richness varied significantly across the range according to chloroplast and nuclear microsatellites, which pointed to higher diversity in West Africa. A high but contrasted level of differentiation was revealed among populations with a clear phylogeographic signal, with both nuclear (F(ST) = 0.21; R(ST) = 0.28; R(ST) > R(ST) (permuted)) and chloroplast simple sequence repeats (SSRs) (G(ST) = 0.81; N(ST) = 0.90; N(ST) > N(ST) (permuted)). We identified a strong geographically related structure separating western and eastern populations, and a substructure in the eastern part of the area consistent with subspecies distinction. Using ENM, we deduced that perturbations during the LGM fragmented the potential eastern distribution of shea tree, but not its distribution in West Africa. Our main results suggest that climate variations are the major factor explaining the genetic pattern of V. paradoxa.     

Genetic changes associated to declining populations of Formica ants in fragmented forest landscape

We monitored populations of two wood ant species, Formica aquilonia and Formica lugubris, through annual mapping of the colonies in a fragmenting forest landscape from 1966 to 1998. The genetic population structure was studied at the end of the study period by using 12 microsatellite loci. Fragmentation of forest led to a decline and spatial redistribution of populations. Changes in the spatial distribution were particularly pronounced in the highly polygynous (many queens in a single nest) species F. aquilonia, whose local populations declined or became extinct, or relocated themselves and colonized new patches. The genetic relationships of the remaining subpopulations indicated the historical developments, revealing the boundaries of the historical populations (high values of genetic differentiation, F(ST)), recolonization histories (genetic affinities revealed by Bayesian analyses) and population decline (reduced variation). Big genetic differences could be detected over short distances, so differentiation also depended on social factors. Our results showed that a genetic study can be reliably used to dissect the recent historical changes underlying the present population structure, and that species with different social structures can respond differently to habitat changes. Combining our demographic and genetic results suggests that habitat fragmentation forms a clear threat on a local scale with large negative effects on ant population viability.     

Surnames in the Azores: analysis of the isonymy structure

Geographic isolation is a significant factor to consider when characterizing human populations. The knowledge of the genetic structure of isolated populations has been of great importance to disease-locus positioning and gene identification. To investigate the genetic structure of the Azorean population, we conducted a survey based on the frequencies of surnames listed in the 2001 telephone book. We calculated the following parameters: isonymy (I), the random component of inbreeding (F(ST)), genetic diversity according to Fisher (alpha), Karlin-McGregor's migration rate (v), and Nei's distance. For the 1,271 subscribers and 163 different surnames, Graciosa island presented the lowest value of abundance of surnames (alpha = 15.75), suggesting great genetic isolation compared to the other eight islands. Migration, calculated on the basis of the diversity of surnames within islands, ranged from 0.2747 (Corvo island) to 0.0026 (S?o Miguel island), indicating that people migrated preferentially toward the economically more developed islands. The value of the random component of inbreeding obtained for the whole population (F(ST) = 0.0039) indicates little genetic differentiation (Wright's F(ST) < 0.05). Moreover, isonymy similarity revealed using the UPGMA method shows three subclusters corresponding to the geographic distribution of the islands.     

Population genetic structure of Anopheles arabiensis mosquitoes in Ethiopia and Eritrea

This study examined the population genetic structure of the major malaria vector, Anopheles arabiensis mosquitoes, in Ethiopia and Eritrea. Ethiopia and Eritrea have great geographical diversity, with high mountains, rugged plateaus, deep gorges, and rolling plains. The plateau is bisected diagonally by the Great Rift Valley into the Northwestern Highlands and the Southeastern Highlands. Five A. arabiensis populations from the Northwestern Highlands region and two populations from high-altitude sites in the Great Rift Valley were genotyped using six microsatellite markers to estimate the genetic diversity and population genetic structure of A. arabiensis. We found that A. arabiensis populations from the Northwestern Highlands and the Great Rift Valley region showed a similar level of genetic diversity. The genetic differentiation (F(ST)) of the five mosquito populations within the Northwestern Highlands region was 0.038 (P <.001), while the two populations within the Great Rift Valley showed little genetic differentiation (F(ST) = 0.007, P <.01). The degree of genetic differentiation between the Northwestern Highlands region and the Great Rift Valley region was small but statistically significant (F(ST) = 0.017, P <.001). The population genetic structure of A. arabiensis in the study area did not follow the isolation-by-distance model (r(2) = 0.014, P >.05). The low F(ST) estimates for A. arabiensis populations in Ethiopia and Eritrea are consistent with the general population genetic structure of this species in East Africa based on other molecular markers.     

Phylogeography of the montane caddisfly Drusus discolor: evidence for multiple refugia and periglacial survival

We studied the genetic population structure and phylogeography of the montane caddisfly Drusus discolor across its entire range in central and southern Europe. The species is restricted to mountain regions and exhibits an insular distribution across the major mountain ranges. Mitochondrial sequence data (COI) of 254 individuals from the entire species range is analysed to reveal population genetic structure. The data show little molecular variation within populations and regions, but distinct genetic differentiation between mountain ranges. Most populations are significantly differentiated based on F(ST) and exact tests of population differentiation and most haplotypes are unique to a single mountain range. Phylogenetic analyses reveal deep divergence between geographically isolated lineages. Combined, these results suggest that past fragmentation is the prominent process structuring the populations across Europe. We use tests of selective neutrality and mismatch distributions, to study the demographic population history of regions with haplotype overlap. The high level of genetic differentiation between mountain ranges and estimates of demographic history provide evidence for the existence of multiple glacial refugia, including several in central Europe. The study shows that these aquatic organisms reacted differently to Pleistocene cooling than many terrestrial species. They persisted in numerous refugia over multiple glacial cycles, allowing many local endemic clades to form.     

SGS--Spatial Genetic Software: a computer program for analysis of spatial genetic and phenotypic structures of individuals and populations.

We have developed a program called Spatial Genetic Software (SGS), which provides a user-friendly Windows tool to analyze both local and broad scale genetic and phenotypic structure. It can deal with nearly any type of genetic data, codominant (allozyme, PCR-RFLP, microsatellite) or dominant (RAPD, AFLP) markers, or biparentally (nuclear) or uniparentally (cpDNA and mtDNA) inherited markers. Data based on any of these markers can be analyzed, either as individual genotypes within a single population (local scale) or as allele or haplotype frequencies from different populations (broad scale). We also include a simple approach to analysis of spatial structure for continuous quantitative traits. The program implements various parameters to analyze spatial genetic and phenotypic structure: Moran's index, Geary's index, number of alleles in common, and approaches using genetic distances and F(ST) values. The statistical significance of all measures is verified by the use of a permutation test. The results are assessed by graphics that can be integrated, via the clipboard, to other Windows programs. The details of the computations are given in a table and can be stored as ASCII files.     

Population structure and demographic inferences concerning the endangered onychophoran species Epiperipatus acacioi (Onychophora: Peripatidae)

Epiperipatus acacioi (Onychophora: Peripatidae) is an endemic species of the Atlantic rainforest in southeastern Brazil, with a restricted known distribution, found only in two nearby areas (Tripuí and Itacolomi). Mitochondrial gene COI sequences of 93 specimens collected across the known range of E. acacioi were used to assess the extant genetic diversity and patterns of genetic structure, as well as to infer the demographic history of this species. We found considerable variability within the populations, even though there has been recent environmental disturbance in these habitats. The samples from the two areas where this species is found showed significantly different COI sequences and constitute two distinct populations [exact test of sample differentiation (P = 0.0008) and pairwise F(ST) analyses (F(ST) = 0.214, P < 0.00001)]. However, there was little genetic differentiation among samples from different sampling sites within populations, suggesting that the potential for dispersal of E. acacioi greater than would have been expected, based on their cryptic behavior and reduced vagility. Mismatch analyses and neutrality tests revealed evidence of recent population expansion processes for both populations, possibly related to variations in the past distribution of this species.     

Experimental demonstration of a causal relationship between heterogeneity of selection and genetic differentiation in quantitative traits

Comparisons of estimates of genetic differentiation at molecular markers (F(ST)) and at quantitative traits (Q(ST)) are a means of inferring the level and heterogeneity of selection in natural populations. However, such comparisons are questionable because they require that the influence of drift and selection on Q(ST) be detectable over possible background influences of environmental or nonadditive genetic effects on Q(ST)-values. Here we test this using an experimental evolution approach in metapopulations of Arabidopsis thaliana experiencing different levels of drift and selection heterogeneity. We estimated the intensity and heterogeneity of selection on morphological and phenological traits via selection differentials. We demonstrate that Q(ST)-values increased with increasing selection heterogeneity when genetic drift was limited. The effect of selection on Q(ST) was thus detectable despite significant genotype-by-environment interactions that most probably biased the estimates of genetic differentiation. Although they cannot be used as a direct validation of the conclusions of prior studies, our results strongly support both the relevance of Q(ST) as an estimator of genetic differentiation and the role of local selection in shaping the genetic differentiation of natural populations.